01 Installation Configuration

Installation

Simply get the MKP from out github or exchange.checkmk.com and insall it either wie the "Extension Mananger" or the mkp-tool on the CLI as usual. Not other dependencies are required.

Configuration

• After the installation you can configure the plugin like a normal active check.
• Go to setup an search for "Automated downtimes"

The create rule which looks like this:

Be sure also to use the inline-help!

1 Like with any activecheck give the service to be created a name

2 Automated downtime monitor

This section defines what host/services should be monitored and on what we should react on (donwtimes, service-output, host/services) state:

Monitor a host

Here you can enter the name of a host-object to watch. You can use macros like $HOSTNAME$ here.

Monitor a service

Here you can specify the name of a host-object and a service and a regex for matching the service-output. So if the service-output of (at least one) of the specified service matches, Downtimes will be set on the dependencies.

3 Dependency detection

Here we define how dependencies (hosts and/or services) are detected. The found dependencies will set into downtime, when the "Maintenance monitor" determines so. (Note: When referring to THISHOST: THISHOST is the host the active check is bound to)

Those strategies are available:

Fully automated

In this strategy

• child-hosts (via parent-child definition),
• hosts having THISHOST in their name and
• services with THISHOST in their description will be included.

Usually used with Monitor Hostname == $HOSTNAME$

This also works recursivly if you let the plugin run on all hosts: So a you manually set a downtime on Gateway "MUC" will triggle down like this

MUC
+--- CoreSW/Interface MUC
+--- MUC-SRV01
     +
     +--- CoreSW/Interface MUC-SRV01
     +--- MUC-SRV01-IDRAC
          +
          +--- CoreSW/Interface MUC-SRV01-IDRAC

Manual selection

A manual selection of hosts and/or service can be specified via RegExes. If you want a host to be put into downtime leave the service-regex empty.

Search options

This options influncence the behavior when seaching for dependencies. See inline help for details

4 Default downtimes in minutes

• Should be at minimum 2.5*x of the normal service check interval.
• The lower the interval, the more often the plugin has to renew then downtimes
• Downtimes get canceled if they are no longer used. So the only downside of a longer downtime is in the case the plugin get disabled. Then you have to wait longer until the downtimes expire on their on (on you have to manually remove them)

5 Conection options

Automation user

Required must have enough rights to see all hosts/services. Used to connect to LQ+RestAPIS

Connect to this central site

In case of a distributed setup you must define the url+ ports on how to connect to the central site. Otherwise only the local site will be queried.

You can only use this to speed up REST-API call in local setups by defining 127.0.0.1 and the 5xxx-TCPPort for your instance. (This bypasses the SSL-Proxy in front of the instance)

(!) This option must be also specified, when you are running in docker, or your frontend-apache is not reachable via localhost. In this case also specify 127.0.0.1 and the respective 5xxx-Backend-Port

6 Servicetimeout

In larger environments the rebuilding of the cache may take longer than the default serivcetimeout of 60s. so a larger value must be allow via "Service Timeout"-rule im CEE or ~/etc/nagios on CRE.